Automotive and many industrial lubricating oils are usually formulated from paraffin based petroleum distillate oils or from synthetic base lubricating oils. Lubricating oils are combined with additives such as soaps, extreme pressure (E.P.) agents, viscosity index (V.I.) improvers, antifoamants, rust inhibitors, antiwear agents, antioxidants, and polymeric dispersants to produce an engine lubricating oil of SAE 5 to SAE 60 viscosity.
After use, this oil is collected from truck and bus fleets, automobile service facilities, municipal motor oil recycling centers and retail stores. There is also a significant volume of oil collected from the industrial sector, e.g., cutting, stamping and coolant oils, which is collected on a direct basis or is collected from oily-water dehydrating facilities. This collected oil contains organo-metallic additives such as zinc dialkylthiophosphate from the original lubricating oil formulation, sludge formed in the engine, and water. The used oil may also contain contaminants such as waste grease, brake fluid, transmission oil, transformer oil, railroad lubricant, crude oil, antifreeze, dry cleaning fluid, degreasing solvents such as trichloroethylene, edible fats and oils, mineral acids, soot, earth and waste of unknown origin.
Reclaiming of waste oil is largely carried out by small processors using various processes tailored to the available waste oil, product demands, and local environmental considerations. Such processes at a minimum include partial de-watering and coarse filtering. Some more sophisticated processors may practice chemical demetallizing or distillation. The presence of organo-metallics in waste oils such as zinc dialkylthiophosphate results in decomposition of the zinc dialkyldithiophospnate to form a carbonaceous layer rich in zinc and often other metals such as calcium, magnesium and other metals present as additives and thus difficult if not impossible to process. The carbonaceous layer containing the various metals forms rapidly on heated surfaces and can develop to a thickness of more than 1 mm in 24 hours. This layer not only reduces the heat transfer coefficient of tubular heaters rapidly, it also results in substantial or total occlusion of these tubes within a few days.
Successful reclaiming processes require the reduction of the organo-metallics (or ash) content to a level at which the hot oil does not foul heated surfaces. Such reduction can be carried out by chemical processes which include reacting cation phosphate or cation sulfate with the chemically bonded metal to form metallic phosphate or metallic sulfate. U.S. Pat. No. 4,432,865 to Norman, the contents of which are incorporated herein by reference, discloses contacting used motor oil with polyfunctional mineral acid and polyhydroxy compound to react with undesired contaminants to form easily removable reaction products. These chemical processes suffer from attendant disposal problems depending on the metal by-products formed.
Ash content can also be reduced by heating the used lubricating oil to decompose the organo-metallic additives. However, indirect heat exchange surfaces cannot be maintained above 400.degree. F. (204.degree. C.) for extended periods without extensive fouling and deposition of metals from the additives. Used lubricating oils can be heated to an additive decomposition temperature of 400.degree. F. (204.degree. C.) to 1000.degree. F. (538.degree. C.) by direct heat exchange by mixing with a heated product oil as disclosed in U.S. Pat. No. 5,447,628 to Harrison, et al., the contents of which are incorporated herein by reference. However, dilution of the product oil with used oil requires reprocessing already processed product oil . . .
UOP's Hy-Lube process, described in U.S. Pat. Nos. 5,244,565 and 5,302,282, and many more, uses a hot circulating hydrogen stream as a heating medium to avoid deposition of decomposed organo-metallic compounds on heating surfaces.
The problem of fouling of heated surfaces can be ameliorated to some extent by gentler heating. Some processes, such as the fixed bed version of catalytic cracking, the Houdry process, used a molten salt bath to provide controlled, somewhat gentle heating of vaporized liquid hydrocarbon passing through tubes of catalyst immersed in the salt bath. Molten metal baths have also been used as a convenient way to heat difficult to processes substances to a control temperature, e.g., flammability of some plastics is tested by putting a flask with plastic into a bath of molten metal. Use of molten salt bath, or molten metal bath, or condensing high temperature vapor, could be used to reduce uneven heating of heat exchange surface and thereby reduce dT across a metal surface and perhaps slow the fouling of metal surfaces in ULO service, but the additives in the ULO would still tend to decompose on the hottest surface, which would be the heat exchanger tubes.
Although not related to ULO heating, in addition to the use of molten metal or molten salt for indirect heating as discussed above, there has been use, either commercial, or reported in the patent literature, of use of molten metal for direct contact heating of various substances. The float process for making glass is almost 50 years old. Molten metal, primarily lead, for heating coal or shale has been practiced in one form or another for almost 100 years. There are recent reports and patents on use of molten metal baths for waste pyrolysis, and conversion of latex, by heating ground up plants in a metal bath to make an oily overhead product. Also somewhat related, but even more different than anything discussed above, is the HyMelt® process, using molten iron beds for dissolution of various feed stocks. Temperatures in the HyMelt process are so high that if a liquid hydrocarbon feed is fed to a HyMelt reactor, the feed almost instantaneously dissociates in hydrogen and carbon, with the carbon dissolving in the molten iron. This is an excellent process for dissociating a hydrocarbon into its elemental constituents, but may be overkill for, e.g., reprocessing ULO, when all that is needed is enough heating to vaporize the lube boiling range components.
Extensive work has also been done on use of molten salt baths to oxidize unwanted and difficult to process streams. Usually the salt baths are heat sources and reagents, i.e., intended to react with the feed, as reported in U.S. Pat. No. 3,845,910 or 4,602,574, which are incorporated by reference.
Some researchers took the position that fouling of metal surfaces during ULO processing was going to happen, and that the best way to deal with it was to inject something into the ULO which would scrub the metal clean, i.e., injecting an abrasive material.
Solvent extraction with light paraffin solvents such as propane, butane, pentane and mixtures thereof have been practiced by Interline and others. Details of the Interline Process are provided in U.S. Pat. No. 5,286,380 and U.S. Pat. No. 5,556,548. While the extraction approach seems like an elegant solution to the problem of processing ULO, the process may be relatively expensive to operate. Their quarterly report of May 15, 2002, reports that “It has become evident that demanding royalties based on production is impractical in many situations and countries. Unless and until the re-refined oil produced in a plant can be sold at prices comparable to base lubricating oils, collecting royalties based on production will be difficult. This reality was experienced in Korea, where the royalty was terminated for the first plant, and in England where the royalties were reduced and deferred until the plant becomes profitable.
A breakthrough in ULO processing occurred with direct contact heating of the ULO with steam or a non-hydrogenating gas. This approach solved the problem of zinc additive decomposition fouling of hot metal surfaces, by ensuring that the metal surfaces holding the ULO were always relatively cool. The hottest spot in these ULO process was the point of vapor injection. Decomposing additives had only themselves to condense upon. Such a vapor injection ULO process was disclosed in my earlier patent, U.S. Pat. No. 6,068,759, Process for Recovering Lube Oil Base Stocks from Used Motor Oil . . . and in U.S. Pat. No. 6,447,672, Continuous Plural State Heated Vapor Injection Process for Recovering Lube Oil Base Stocks from Used Motor Oil . . . Other variations on the theme of ULO vapor injection processes are disclosed in U.S. Pat. No. 6,402,937 Pumped Recycle Vapor and U.S. Pat. No. 6,402,938, Vaporization of Used Motor Oil with Non-hydrogenating Recycle Vapor, which are incorporated by reference.
The “state of the art” of used motor oil processing could be summarized as follows:
Chemical additive and extraction approaches can be used to react with, or extract everything but, zinc additives, but costs associated with such processes are apparently high, as evidenced by little commercial use. Additives could be extracted, but the operating costs are high.
Indirect heating, in a fired heater, causes rapid fouling of metal surfaces. Using milder heating, via a double boiler approach or molten metal heating medium, can minimize but not eliminate fouling on hot metal surfaces.
Direct contact heating with high pressure hydrogen may eliminate fouling but requires high capital and operating expenses.
Direct contact heating, with recycled product oil, helps but requires processing the ULO twice.
Oxidation, either by burning as a low grade fuel, or perhaps as part of a salt bath oxidation process for waste streams.
Direct contact heating with steam or non-hydrogenating vapor, as reported in my U.S. Pat. No. 6,068,759 and the related patents discussed above, is believed to be the best available technology. This approach requires only moderate capital investment and moderate operating expense when steam is the injected vapor, but the process can create a water disposal problem and is thermally less efficient because the latent heat of water is lost when the steam is condensed against cooling water or air in a heat exchanger. When other vapors are injected for heating e.g., propane, the water problem goes away but large volumes of vapor are needed to provide sufficient heat input, so costs increase to heat and recycle such vapor streams.
Although my earlier work, steam injection for direct contact heat exchange, solved the worst problem, fouling on hot metal surfaces, it had some deficiencies as briefly noted above. I wanted an even better approach.
I thought about steam injection. The steam injection process seemed nice and simple, because it was easy to heat water to make steam. Unfortunately, using large amounts of water created a potential water disposal problem and produced a relatively “wet” plant, with many potential areas for corrosion as the steam condensed. Re-using the condensed water was possible, but there are concerns about the amount of water treatment required to remove chlorides, etc, so that corrosion and/or plugging of the tubes in the fired heater would not be a problem. Large volumes of steam were required, which resulted in relatively large plant volumes, at least until some or all of the injected steam was condensed. I realized that although the use of steam was a great advance in the art, it might not be the best approach.
The “pumped vapor” approach, use of propane or other recycle hydrocarbon vapor eliminates many concerns about water, but required a more complicated plant to recycle the hydrocarbon vapor. Large molar volumes of injected vapor are needed because of the relatively low heat capacity of hydrocarbon vapors. Condensation and separation of multiple hydrocarbon species, both injected heating vapors and recovered lubricating components, is more complicated than cooling everything and allowing water and oil to separate as separate phases.
I wanted to retain the beneficial features of heating the ULO by injecting something hot into it, but avoid the problems created by using either steam or a light hydrocarbon vapor as the heating medium. I found a way to overcome these deficiencies, by using a non-pyrolizing molten salt bath as the heating fluid.
There are many salts available which are fluid at relatively low temperatures which have ideal properties for use herein. They are relatively non-corrosive, especially when used in a reducing atmosphere. They are inexpensive and easy to contain. Molten salt is sufficiently dense to hold a lot of heat, permitting reasonably efficient heating of waste streams. They are not volatile, so they do not contribute to air or water pollution. They are immiscible with ULO so the decomposition products and trash found in the ULO can be easily removed from the molten salt bath. Molten salt also permits a flexible design approach, permitting injection of the molten salt into the oil or vice versa, though not necessarily with equivalent results. When oil is injected into a molten salt bath, it is easy to increase or decrease process severity by changing the depth of molten salt in the bath or the temperature of the salt or the pressure in the molten salt bath.